Relaxation Processes of Molecules on a Solution Surface Studied by

Nov 10, 1998 - Cluster Research Laboratory, Toyota Technological Institute: in East ... Genesis Research Institute, Inc., 717-86 Futamata, Ichikawa, C...
1 downloads 0 Views 75KB Size
9540

J. Phys. Chem. B 1999, 103, 9540-9544

Relaxation Processes of Molecules on a Solution Surface Studied by the Two-Color Pump-Probe Ionization Technique Noriko Horimoto,† Fumitaka Mafune´ , and Tamotsu Kondow* Cluster Research Laboratory, Toyota Technological Institute: in East Tokyo Laboratory, Genesis Research Institute, Inc., 717-86 Futamata, Ichikawa, Chiba 272-0001, Japan ReceiVed: NoVember 10, 1998

A propanol (PrOH) solution of aniline (AN) was introduced into a vacuum chamber as a continuous liquid flow (liquid beam) and was irradiated with pump and probe laser beams having wavelengths of 266 and 500-600 nm, respectively. Ions produced by multiphoton absorption and ejected from the liquid beam surface were analyzed by a time-of-flight mass spectrometer. The mass spectra show that AN+(PrOH)n and H+(PrOH)n are produced. The AN+(PrOH)n and H+(PrOH)n intensities were measured as a function of the delay time between the pump and the probe lasers. The lifetime for relaxation of an excited aniline molecule on the liquid surface was obtained to be 7.8 ns from the dependence of the AN+(PrOH)n intensity on the delay time. In comparison with the lifetimes of an isolated aniline molecule in the gas phase and an aniline molecule in bulk solution, it is concluded that the aniline molecule does not strongly interact with its surrounding propanol molecules on the surface. Variation of the H+(PrOH)n intensity with the delay time indicates that H+ produced by dissociative ionization of an aniline molecule is ejected with several propanol molecules as H+(PrOH)n.

Introduction A solute molecule on a solution surface is not completely solvated with its solvent molecules and is subjected to an asymmetric field so that the solvation structures and dynamics are very different from those inside the solution. The specificity of the solution surface, in turn, provides a unique opportunity to investigate how the solute molecule behaves under the asymmetric field built by the solvent molecules on the surface. Several spectroscopic techniques have been applied for the studies of surface molecules. For example, on the surface of a benzyl alcohol-formamide liquid mixture, benzyl alcohol is found to be enriched as expected from several macroscopic measurements such as surface tension.1-3 The kinetic energy distributions of electrons produced by photon and metastable atom impact, having different ionization depths, have revealed that benzyl alcohol tends to be segregated on the surface of a benzyl alcohol-formamide mixture even when they are homogeneously dissolved in the bulk.1 A sudden change of the surface composition in a water-formamide mixture with a slight change of the bulk composition has been observed by photoelectron spectroscopy:2 The change is attributed to a solvent-solvate transition.2,4 It has also been clarified in this study2 that it takes 50-60 µs to reach an equilibrium value of the surface composition after changing the bulk composition. The latter information is related to the dynamical behavior of the solution surface in the long time regime and provided insight into the large-scale motion of the surface. Furthermore, the dynamics of the molecules on a liquid surface (molecular motion, chemical reaction, etc.) has been investigated by several methods. Scattering of atoms and molecules from a liquid surface has revealed a microscopic picture of how impinging atoms and molecules are accommodated in or rebounded from the sur* To whom correspondence should be addressed. E-mail: kondow@ mail.cluster-unet.ocn.ne.jp. † Permanent address: The Institute of Physical and Chemical Research (RIKEN), 2-1 Hirosawa, Wako, Saitama 351-0198, Japan.

face.5,6 The orientational relaxation time of rhodamine 6G on an aqueous solution surface has been observed to be 400-500 ps by second harmonic generation. This relaxation time is longer than the rotational relaxation time of 200 ps in the bulk. This finding is explained by a large frictional resistance exerted on the surface rather than on the bulk.7 Ion-molecule reactions induced by multiphoton excitation on various liquid surfaces have been investigated by using liquid-beam multiphoton ionization mass spectrometry.8-18 In addition to the studies of the surface dynamics, it is important to obtain information on energy redistribution accompanied by the surface dynamic processes. A decay rate measurement of an excited surface molecule on a liquid surface is one of the most suitable means to gain an insight into the energy flow in relation to the local environment that surrounds the excited molecule. A system in which the energy redistribution depends sensitively on the local solvent structure should be suitable for investigation. For example, the electronic relaxation of an excited aniline molecule in an alcohol solution is considered to be shortened by the hydrogen bonding between aniline and alcohol molecules.19 A combination of pump-probe two-color photoionization with the liquid beam technique enables the measurement of the decay rate on the liquid surface. In addition to the usual merit of the liquid beam technique that ions produced in the vicinity of a liquid surface can be detected selectively by mass spectrometry, the time dependence of transient excited species on a liquid surface can be observed by introduction of the two-color photoionization. We performed a decay measurement of an excited aniline molecule (AN) on the surface of a 1-propanol (PrOH) solution by using the method mentioned above. It is found that the lifetime of an aniline molecule in the S1 state on the solution surface is comparable to the lifetime of an isolated aniline molecule in the S1 state. This finding is related to the weak interaction between the aniline molecule and its surrounding propanol molecules on the surface. There exists a decay channel of the

10.1021/jp984384f CCC: $18.00 © 1999 American Chemical Society Published on Web 10/14/1999

Relaxation Processes of Molecules on a Solution Surface

J. Phys. Chem. B, Vol. 103, No. 44, 1999 9541 SCHEME 1

Figure 1. Schematic diagram of the liquid beam apparatus.

aniline molecule in the S1 state, which correlates with the formation of H+(PrOH)n. The cluster ion, H+(PrOH)n, is considered to be formed via the S1 and T2 states of an aniline molecule. Experimental Section Figure 1 shows a schematic diagram of the experimental apparatus used in the present study, which consists of a liquid beam source and a time-of-flight (TOF) mass spectrometer. A continuous laminar liquid flow (liquid beam) was introduced downward through an aperture with a diameter of 20 µm into a vacuum chamber by applying a pressure of ∼20 atm with a Shimadzu LC-6A pump. The flow rate was kept constant at 0.2 mL/min. Aniline and 1-propanol (Kanto Chemical Co., Inc.; G-grade) were used without further purification. The liquid beam was captured by a liquid N2 trap (hatched areas in Figure 1) 10 cm downstream from the aperture. The chamber was further evacuated down to 10-5-10-6 Torr by a 1200 L/s diffusion pump. In this pressure range, the mean-free path of the molecules in the gas phase exceeds 1 m so that molecules do not stagnate in the close vicinity of the solution surface. This feature is supported by the fact that electrons from a liquid surface have been observed after metastable-atom impact,20 where the ionization probability of a molecule colliding with a metastable atom is almost unity. Evidently, there is no substantial gas layer on the liquid beam surface because the metastable atoms reach the surface without suffering collisional deactivation by molecules in the gas phase. Traveling a distance of 1 mm from the aperture, the liquid beam was crossed with two laser beams (pump and probe lasers) at the first acceleration region of the TOF mass spectrometer. A UV laser at 266 nm was used as the pump laser, which is resonant with the S0-S1 transition of an aniline molecule; the laser was provided by the fourth harmonic of a Quanta-ray GCR-3 Nd:YAG laser. The probe laser for ionization was provided by a Spectra-Physics MOPO optical parametric oscillator (500-600 nm). The pulse width of the lasers was ∼5 ns. The delay time between the pump and probe lasers was controlled by a Stanford Research Systems DG-535 digital delay pulse generator. The pump and the probe lasers were focused onto the liquid beam by a lens with a focal length of 400 mm. Throughout this study, the liquid beam was completely within the pump and probe laser spots. The method for measuring the lifetime of the S1 state is depicted in Scheme 1. Ions ejected from the liquid beam were accelerated by a pulsed electric field in the first acceleration region of the TOF mass spectrometer, in the direction perpendicular to both the liquid and the laser beams. A delay time from the ionization to

the ion extraction was varied in the range of 0-5 µs so as to improve the mass resolution. The ions were then steered and focused by a set of vertical and horizontal deflectors and an einzel lens. After traversing a 1-m field free region, the ions were detected by a Murata EMS-6081B Ceratron electron multiplier. Signals from the multiplier were amplified and processed by a Yokogawa DL 1200E transient digitizer based on an NEC 9801 microcomputer. The mass resolution, m/∆m, was found to be more than 85 at m ) 200 in the present experimental conditions. In the present study, the intensities of an aniline ion and its fragment ions were found to be much higher than those of the cluster ions of interest. In order to observe cluster ions in an effective manner, a mass gate was mounted in the field-free region of the TOF mass spectrometer to select the ions of interest. Rejecting the undesired ions, namely aniline ion and its fragment ions in the present study, improved the detector sensitivity. The mass gate consisted of two sets of horizontal deflectors which were mounted serially with a gap of 2 cm between them. A metal grid electrically grounded was placed between the two deflectors. A pulsed electric field was applied to the mass gate so that only the ions having masses larger than that of an aniline ion were admitted. Results Figure 2 shows typical TOF mass spectra of ions produced from a 0.5 M aniline solution in propanol. Panel a shows the mass spectrum of ions produced by irradiation of only the pump laser at 266 nm. A series of ion peaks (m/z ) 93, 153, 213, 273, ..., and m/z ) 61, 121, 181, ...) observed in the spectrum are assigned to AN+(PrOH)n and H+(PrOH)n, respectively. These ions are produced by two-photon absorption of the pump laser. Panel b shows the mass spectrum of the ions when the probe laser at 500 nm was introduced simultaneously with the pump laser. Although the mass spectral feature does not change significantly, the intensity of AN+(PrOH)n increases by the addition of the probe laser: Note that no cluster ions were observed in the mass spectrum as long as the pump laser irradiates outside the liquid beam, even if the probe laser crosses the liquid beam or illuminates outside the liquid beam. The increment in the intensities of AN+(PrOH)n and H+(PrOH)n by the addition of the probe laser gives rise to the spectrum (the differential spectrum, panel c, obtained by subtracting a from b) of the ions produced only by the two-color two-photon pump-probe photoionization. The total increment of the X+(PrOH)n (X ) AN or H) intensity caused by addition of the probe laser to the pump laser, , is defined as ∆I(total) X

∆I(total) ) X

∑n ∆I(n)X

(1)

9542 J. Phys. Chem. B, Vol. 103, No. 44, 1999

Horimoto et al.

Figure 4. Total increase of the abundances of AN+(PrOH)n (n g 2), ∆Itotal AN , plotted against the pump-probe delay time (0-35 ns). The relaxation time of an excited aniline molecule on the solution surface is measured to be 7.8 ( 0.8 ns.

Figure 2. Mass spectrum of ions produced from a 0.5 M aniline solution of 1-propanol under the irradiation of the pump laser at 266 nm (a) and the pump and the probe laser at 500 nm (b). A subtracted spectrum (b-a) is exhibited in c. The two lasers illuminate the liquid beam simultaneously (delay time 0 ns). Solvated aniline cluster ions, AN+(PrOH)n (m/z ) 93, 153, 213, 273, ...), increase in intensity by addition of the probe laser, while the protonated propanol cluster ions, H+(PrOH)n, scarcely change.

Figure 5. Increase of the abundances of all the H+(PrOH)n with n g 3, ∆Itotal H , plotted against the pump-probe delay time (0-50 ns). Note that ∆I(total) is negligibly small at delay time ) 0 but shows a sharp H increase within the first 5 ns of delay time.

Figure 3. Intensity of AN+(PrOH)n (n g 2), plotted against the power of the laser. (a) Pump laser power dependence (probe laser power 0.8 mJ/pulse). (b) Probe laser power dependence (pump laser power 170 µJ/pulse). + where ∆I(n) X stands for an increment of X (PrOH)n. The contributions of X+(PrOH)n with n < 2 for X ) AN and n < 3 for X ) H are excluded from the sum because these ions are not well separated from AN+ due to insufficient mass resolution of the mass gate. This treatment is warranted because an ion, X+(PrOH)n, with a different n was found to depend almost identically on the laser power and the delay time. Panels a and b of Figure 3 represent ∆I(total) AN plotted against the power of the pump and the probe lasers, respectively. The value, ∆I(total) AN , depends linearly on both the power of the pump and the probe lasers. It was also found that ∆I(total) depends H linearly both on the power of the pump and the probe lasers. Figure 4 shows ∆I(total) as a function of the pump-probe AN delay time (0-35 ns). As the delay time increases, ∆I(total) AN decreases exponentially, with the characteristic lifetime of 7.8 ( 0.8 ns, where the pulse widths of the two lasers are deconvoluted from the time profile of ∆I(total) AN . At negative delay times, ∆I(total) decreased drastically and became half AN

within